Facility level variation in the adoption of neo-adjuvant chemotherapy is associated with higher surgical morbidity for patients with advanced stage epithelial ovarian carcinoma (558).

Objectives: To investigate outcomes of patients with advanced-stage epithelial ovarian cancer (EOC) based on facility-level adoption of neoadjuvant chemotherapy (NACT). Methods: The National Cancer Database was accessed, and patients diagnosed between 2010-2016 with stage III-IV EOC at the reporting facility were identified. For facilities that reported at least 120 patients, the percentage of those managed with NACT (with or without IDS) was calculated. Facilities with the lowest and highest quartile of NACT rate comprised the low and high-adopting groups, respectively. Clinical-pathologic characteristics were compared with the Chi-square test. For patients with available follow-up, median overall survival was compared with the log-rank test, while a Cox model was constructed to control for confounders. Results: A total of 71 facilities reported at least 120 patients with advanced-stage EOC; the mean percentage managed with NACT was 39.29% (range: 5.26%-64.79%). High and low-adopting facilities managed ≥47.65% and ≤32.76% of patients with NACT. A total of 6954 patients were managed at these facilities; 2982 (42.9%) and 3972 (57.1%) at low and high-adopting facilities, respectively. Patients managed in high-adopting facilities were more likely to have stage IV (31.1% vs 23.3%, p<0.001), and comorbid conditions (19.7% vs 16.3%, p<0.001) while age was comparable (median 64 vs 63 yrs, p=0.09). Low adopting facilities were more likely to perform PDS for patients with stage IV (54.4% vs 24.5%, p<0.001) disease, or those with comorbid conditions (71.5% vs 40.5%, p<0.001). For patients undergoing surgery (either PDS or IDS) (n =6043), 90-day mortality (6.9% vs 2.8%, p<0.001), unplanned 30-day re-admission (6% vs 3.9%, p<0.001) and prolonged hospital stay rate (defined as >10 days) (12.9% vs 10%, p<0.001) were higher among patients managed at low-adopting facilities. The rate of complete gross resection (CGR) was higher among patients managed at a high-adopting facility (55.4% vs 47.5%, p<0.001), while they were less likely to undergo bowel resection (28.1% vs 35.9%, p<0.001). Overall, patients who underwent PDS had better OS compared to those who had NACT (median 42.84 vs 27.86 ms, p<0.001). Median OS for patients managed in high (n =3437) and low-adopting (n =2581) facilities was 35.42 and 36.4 months, respectively (p=0.98). After controlling for confounders, those managed at high-adopting facilities had a trend towards better OS (HR: 0.92, 95% CI: 0.85-0.99). For patients undergoing PDS, those managed at a high- adopting facility (n =1601) had better OS compared to those managed at a low-adopting facility (n =1991) (median 48.46 vs 39.36 months, p<0.001). For patients receiving NACT, OS was comparable between high (n =1836) and low-adopting facilities (n =590) (median 28.78 vs 25.13 ms, p=0.081). For patients who underwent surgery (either PDS or IDS), those managed at a high-adopting facility had better OS after controlling for confounders (HR: 0.87, 95% CI: 0.80-0.94). However, when including patients who survived at least 90 days, OS was comparable (HR: 0.94, 95% CI: 0.87-1.02). Conclusions: Facility level utilization of NACT varies substantially among high-volume centers. Despite treating a patient population with overall more advanced disease and more comorbid conditions, high-adopting facilities have lower surgical morbidity and mortality with no detrimental impact on long-term survival. These findings highlight the need to avoid PDS in ovarian cancer patients who, secondary to the extent of disease or comorbid conditions, are unlikely to benefit from the procedure. [ABSTRACT FROM AUTHOR]